Pathogenesis of Dyslunesias in Parlunson’s Disease M. M. Mouradian, MD, I. J. E. Heuser, MD, F. Baronti, MD, G. Fabbrini, MD, J. L. Juncos, MD, and T. N. Chase. MD

Abnormal involuntary movements complicate the man- agement of a majority of patients with advanced Parkin- son’s disease. The ability of levodopa to induce dys- kinesias and alleviate parkinsonism has generally been considered a continuous dose-dependent pharmacolog- ical spectrum. In this study, the acute dose-response profile of intravenously administered levodopa for both inducing dyskinesia and alleviating parkinsonism, and its duration of action on these motor manifestations were evaluated in 52 parkinsonian patients. The minimum dose of levodopa required to produce mild dyskinetic movements was significantly lower in pa- tients with fluctuations in motor response compared with those who had a stable response to standard oral therapy; the minimum dose for antiparkinsonian benefit, however, failed to show significant differences. The rate of disappearance of dyskinetic movements was faster than the rate of reappearance of parkinsonian signs following withdrawal of a steady-state infusion of levodopa. The dissociation of the pharmacodynamic profile of the two major motor effects of levodopa sug- gests their mediation through two . different central pharmacological mechanisms, perhaps involving the two classes of dopamine receptors or other transmitter systems, and could have important implications for the design of future antiparkinsonian agents.

Mouradian MM, Heuser IJE, Baronti F, Fabbrini G, Juncos JL, Chase TN. Pathogenesis of

dyskinesias in Parkinson’s disease. Ann Neurol 1989;25:523-526

Levodopa continues to be the most effective antipar- kinsonian medication available. After several years of therapy, however, management becomes complicated by fluctuations in motor response and abnormal in- voluntary movements. The latter are frequently associ- ated with peak plasma levels of dopa (peak-dose dys- kinesias) and are therefore thought to represent a

From the Experimental Therapeutics Branch, National Institute of Neurological and Communicative Disorders and Stroke, Bethesda, MD. Received Sep 26, 1988, and in revised form Nov 11. Accepted for publication Nov 12, 1988. Address correspondence to Dr Mouradian, NINCDS/NIH, Build- ing 10, Room 5C103, 9000 Rockville Pike, Bethesda, MD 20892.

continuum of the antiparkinsonian action of levodopa based on the same underlying pharmacological mecha- nism. The occurrence of diphasic dyskinesias and the transient coexistence of both parkinsonian signs and choreiform movements in some patients cast doubt on this hypothesis. The presence of different mechanisms could have important implications for the develop- ment of antiparkinsonian medications having a more favorable toxic : therapeutic ratio. To explore the pos- sibility of different mechanisms, the duration of the dyskinetic and antiparkinsonian actions of levodopa and its dose-response profile for both extrapyramidal motor effects were evaluated in parkinsonian patients at various stages of the disease.

Methods Fifty-two patients with idiopathic Parkinson’s disease partici- pated in this study after full disclosure of its purposes, risks, and potential benefits. There were 41 men and 11 women aged 59 * 1 years (mean t SEM) (range, 30 to 76 years). Patients were grouped according to their clinical response to optimal dosage of oral levodopdcarbidopa in one of four categories: (1) patients who had never received levodopa, (2) patients treated with standard levodopa 3 or 4 times per day who had a stable motor response, (3) patients who had wear- ing-off type of fluctuations clearly related to timing of levodopa ingestion every 2 to 3 hours, and (4) patients who had complicated on-off phenomenon that had no apparent relation to timing of drug ingestion or other exogenous fac- tors El-31 (Table). Other antiparkinsonian medications were kept at stable levels throughout the investigation. Not all patients participated in all phases of the study.

Acute levodopa dose response studies [3] were conducted in 39 patients who did or did not have fluctuations (8 levodopa naive, 6 with stable response to standard oral levodopa therapy, 13 with mainly wearing-off phenomenon, and 12 with predominantly on-off phenomenon). Single daily intravenous injections of a wide range of levodopa doses were administered after an overnight fast and with- drawal of all antiparkinsonian medications. Patients were premedicated with 100 mg of carbidopa orally an hour be- fore each bolus.

Duration of levodopa efficacy El, 2] was studied in 32 patients who had fluctuations in motor response and dys- kinesias in response to standard oral therapy (1 3 with wear- ing-off phenomenon and 19 with on-off phenomenon). The drug was infused intravenously at optimal rates for at least 16 hours to ensure steady-state conditions; the infusion was then abruptly stopped and the decline in the motor effects (both parkinsonian and dyskinesia severity) was rated for the next 2 hours. While being infused, patients were medicated with 50 mg of carbidopa orally every 3 hours during waking hours.

Motor ratings for both phases of the study were carried out using a modified Columbia Rating Scale and a modified Abnormal Involuntary Movement Scale (AIMS). Data are presented as means ? standard errors of the means. Phar- macological parameters of antiparkinsonian and dyskinetic

523

Patient Characteristics Compared by Response to Oral hodopa"

Never- Treated Stable Wearing-off On-off

Characteristic Group Responders Group Group

Number of patients 8 6 16 22

Age at onset of symptoms (yr)b 55 2 3 54 ? 5 47 * 2 45 ? 2 Current age (yr) 59 +- 4 62 k 5 58 _t 3 60 _t 2

Duration of symptoms (yr)' 3.4 -t 0.7 8.2 2 1.3 11 t 1.4 15 * 1.0 Staged 1.6 2 0.3 3.2 ? 0.2 3.1 t 0.1 Duration of levodopa treatment (yr)d 0 5.7 ? 1.7 7.7 ? 1.0 12 2 0.8

4.6 ? 0.1

Oral levodopa dose (mg/kg/hr)d 0 0.55 2 0.1 0.64 ? 0.06 1.1 * 0.06

"Values given are the mean ? standard error of mean. bAge of onset in the on-off group was significantly lower than that for the levodopa-naive and stable-responder groups (analysis of variance, p < 0.0001). 'On-off patients had a longer duration of symptoms than any other group. The difference between levodopa-naive and wearing-off groups was also significant (analysis of variance, p < O.OOOI). dNever-treated and on-off groups were different from all other groups (analysis of variance, p < 0.0001). Stage was determined by the Hoehn and Yahr scale measured in the untreated state. Levodopa dose was calculated for waking hours.

actions of levodopa were compared using paired t-tests. The four groups of patients were compared by analysis of vari- ance (ANOVA) followed by Fisher post-hoc test. Intercor- relations between variables were analyzed using Pearson's correlations and by forward linear stepwise regression proce- dures [4}.

Results Threshold doses of levodopa for inducing dyskinesia and alleviating parkinsonism showed differential pat- terns with progression of response groups (Fig 1): The minimum intravenous dose of levodopa required to produce abnormal involuntary movements was signifi- cantly lower in the two groups who had fluctuations (0.68 _t 0.10 mg/kg for the wearing-off group, 0.78 2 0.10 mglkg for the on-off group) than in the two groups who had no instability in motor response (2 1.6 mg/kg for the never-treated group, 1.35 f 0.08 mg/kg for stable responders) (ANOVA, p < 0.0001). In contrast, the smallest dose necessary for a minimally detectable improvement in parkinsonian signs did not differ significantly among the four groups (0.53 2 0.09 mg/kg for the never-treated group, 0.52 & 0.08 mg/kg for stable responders, 0.52 -+ 0.05 mg/kg for the wearing-off group, and 0.7 1 ? 0.09 mg/kg for the on-off group; the differences were not significant). Correlational analysis using stepwise regression re- vealed the threshold dose for dyskinesia to be most clearly correlated with the duration of levodopa treat- ment (r . = -0.596, p < O.OOl), while the threshold dose for antiparkinsonian effects was best related to the overall severity of symptoms ( r = 0.394, p < 0.05).

The duration of the dyskinetic action of levodopa was significantly shorter than that of its antiparkinso- nian effect (Fig 2). The dyskinesia half-time, the time required to diminish AIMS scores by 50% after with-

1.6

1.2

0) Y

2 0.8 E

0.4

n Never Stable Wearing-off On-off

Treated (n=6) (n=13) (n=12) (n=8)

~

Fig I . Threshold doses for the antiparkinsonian ej$ct (open bars) and the dyskinetic action (hatched bars) of levodopa deter- mined from acute intravenous dose-response studies. Data repre- sent means _t standard ewor of means. *p < 0.006; '+p < 0.0001 for difference from dyskinesia threshold.

drawal of a steady-state levodopa infusion, was 25 2 3 minutes; in contrast, the antiparkinsonian efficacy half- time, the time needed for parkinsonian motor scores to achieve 50% of maximum severity seen at the end of the 2-hour observation period, was 50 2 4 minutes (paired t test, p < 0.0001). Moreover, the decay curve for dyskinesia was curvilinear, best described by a qua- dratic function ( R 2 linear = 0.764, R 2 quadratic = 0.979; p < 0.01), while that for parkinsonism was linear ( R 2 linear = 0.990, R2 quadratic = 0.998; not significant).

Discussion Distinctly different pharmacological profiles for dys- kinesias and parkinsonian signs were demonstrated in

524 Annals of Neurology Vol 25 No 5 May 1989

I 30 6 0 ‘ 0 40 80 120

MINUTES

Fig 2. Decay in the antiparkinsonian and dyskinetic ej$cts of levodopa after abrupt withdrawal of a steady-state levodopa in- fusion at optimal doses. Each duta point is the mean k stan- dard error of mean for 32 patients manifsting fEtlctuations in motor response and abnomzal involuntary movements on standard oral therapy. Motor scores at time zero were the mean of three ratings obtained just prior to cessation of infusion and represent a state of optimal therapy for these patients.

this study, with respect to both the threshold dose of levodopa for their manipulation and their decay rates following levodopa withdrawal. These observations are consistent with previous clinical observations that, with time, dyskinesias occur at progressively smaller levo- dopa doses IS} whereas antiparkinsonian efficacy re- mains essentially unchanged [GI. Such discrepancies suggest the mediation of abnormal involuntary move- ments via central mechanisms that are different from those for the antiparkinsonian action of levodopa. These central pharmacodynamic differences presum- ably lie postsynaptic to the dopamine terminals, since one transmitter, dopamine, which is synthesized and stored in the presynaptic pool 171, is primarily respon- sible for both effects. Perhaps the two different, yet complexly interrelated, subclasses of dopamine recep- tors, D-1 and D-2, underlie this clinical pharmacolog- ical discrepancy. Preclinical studies point to the role of the D-1 receptor 181, or the relatively heightened tone of D-1 compared with D-2 receptors in the generation of stereotyped behaviors [9]. Confirmation of this pos- sibility from postmortem or cerebral imaging studies of patients with Parkinson’s disease has not been achieved 110- 12). Nevertheless, the clinical observa- tion that chronic use of the relatively selective D-1 agonist bromocriptine results in a lower incidence of dyskinesias, compared with levodopa 1131, further em- phasizes the potential role of the D-1 receptor in the generation of abnormal involuntary movements. Alter- natively, the different pharmacological mechanisms for the two motor symptoms could involve transmitter

systems parallel with or downstream from the striatal dopamine-responsive cells 114, 151. Furthermore, the nonlinear profile of dyskinesia decay after withdrawal of levodopa may suggest the participation of more than one neurotransmitter in this motor complication.

The present results also help elucidate the cause of dyskinetic movements. The close correlation between threshold dose for levodopa-induced hyperkinetic movements and duration of levodopa therapy suggests that chronic intermittent ingestion of the dopamine precursor might contribute to the appearance of these motor complications. Such periodic stimulation of postsynaptic dopamine receptors is nonphysiological, since the system normally operates tonically under neural control 1161. The relatively increased D-1 re- ceptor binding sites in postmortem brains of patients treated with levodopa compared with those untreated supports this hypothesis C12). However, marked de- generation of the presynaptic dopaminergic cells, which store and control the neuronal release of the transmitter mine, may be a prerequisite for the devel- opment of levodopa-induced dyskinesias, because the latter are essentially limited to patients with Parkin- son’s disease 15, 171.

The potential therapeutic implications of these ob- servations can now be explored. Because of the phar- macological discrepancy of parkinsonism and dys- kinesias, it becomes increasingly difficult to manage patients with advanced Parkinson’s disease using inter- mittent oral levodopa as a monotherapy. Continuous infusion therapies acutely reduce, and may be used for the prophylaxis of, moror fluctuations and d yskinesias 1181. In the meantime, trials of various selective D-2 and D-1 receptor agonists and antagonists in various combinations and ratios might prove of practical value.

Presented in part at the 9th International Symposium on Parkinson’s Disease, Jerusalem, Israel, June 1988.

References 1. Fabbrini G, Mouradian MM, Juncos JL, et al. Levodopa phar-

macokinetic mechanisms and motor response flucruations in Parkinson’s disease. Ann Neurol 1987;2 1:370-376

2. Fabbrini G, Mouradian MM, Juncos JL, et al. Motor fluctuations in Parkinson’s disease: central pathophysiological mechanisms, part I. Ann Neurol 1988;24:366-371

3. Mouradian MM, Juncos JL, Fabbrini G, et al. Motor fluctuations in Parkinson’s disease: central pathophysiological mechanisms, part 11. Ann Neurol 1988;24:372-378

4. Kleinbaum DG, Kupper LL. Applied regression analysis and other multivariable methods. Boston: Duxbury Press, 1978

5. Barbeau A. L-dopa therapy in Parkinson’s disease: a critical re- view of nine years’ experience. Can Med Assoc J 1969;lOl:

6. Cotzias GC, Papavasiliou PS, Gellene R. Modification of parkin- sonism-chronic treatment with L-dopa. N Engl J Med 1969;

791-800

280:337-345

Brief Communication: Mouradian et al: Pathogenesis of Dyskinesias in PD 525

7. Javoy F, Glowinski J. Dynamic characteristics of the “functional compartment” of dopamine in dopaminergic terminals of the rat striatum. J Neurochem 1971;18:1305-1311

8. Groppetti A, Flauto C, Parati E, et al. Dopamine receptor changes in response to prolonged treatment with L-dopa. J Neural Transm 1986(supp1);22:33-45

9. Rosengarten H , Schweitzer JW, Friedhoff AJ. Selective dopa- mine D-2 receptor reduction enhances a D-1 mediated oral dys- kinesia in rats. Life Sci 1986;39:29-35

10. Pimoule C, Shoemaker M, Reynolds GP, Langer SZ. [3H]SCH 23390 labeled D-1 dopamine receptors are unchanged in schizophrenia and Parkinson’s disease. Eur J Pharmacol 1985;

11. Raisman R, Cash R, Ruberg M, et al. Binding of C3H)SCH 23390 to D-1 receptors in the putamen of control and parkinso- nian subjects. Eur J Pharmacol 1985;113:467-468

12. Rinne JO, Rinne JK, Laakso K, et al. Dopamine D-1 receptors in the parkinsonian brain. Brain Res 1985;359:306-310

114~235-237

13. Lees AJ, Stern GM. Sustained bromocriptine therapy in previ- ously untreated patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 198 1;44: 1020-1023

14. Javoy-Agid F, Ruberg M, Taquer H, et al. Biochemical neuro- pathology of Parkinson’s disease. Adv Neurol 1984;40: 189- 197

15. Uhl GR, Hackney GO, Torchia M, et al. Parkinson’s disease: nigral receptor changes support peptidergic role in nigrostriatal modulation. Ann Neurol 1986;20:194-203

16. Bunney BS, Walters JR, Roth RH, Aghajanian GK. Dopamin- ergic neurons: effect of antipsychotic drugs and amphetamine on single cell activity. J Pharmacol Exp Ther 1973;185:560-571

17. Chase TN, Holden EM, Brody JA. Levodopa-induced dys- kinesias. Comparison in parkinsonism-dementia and amyotro- phic lateral sclerosis. Arch Neurol 1973;29:328-330

18. Mouradian MM, Juncos JL, Fabbrini G, Chase TN. Motor fluc- tuations in Parkinson’s disease: pathogenetic and therapeutic studies. Ann Neurol 1987;22:475-479

526 Annals of Neurology Vol 25 N o 5 May 1989